Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133247
Megha Gupta, R. Rama Kishore
Abstract Transmission media is at rise over the internet, due to which the copyright risk is alarming. In such situation, enhanced security is the main concern. Digital image watermarking is the technique that ensures the copyright protection, security, and authenticity of the digital image. This research work recommends a highly secured and robust digital image watermarking system. In this method, the cover image is confused by arbitrarily generated numbers by a six-dimensional hyperchaotic technique, then this permuted image is scattered by applying the Fibonacci Q matrix to generate a scattered host image. This scattered image is decomposed up to 3 levels through DWT and the low pass sub-band caused by DWT are further decomposed by SVD. Singular values generated by SVD are used for watermark embedding as a slight change in the value does not affect the image quality. SVM classifier is used to classify the appropriate location to insert the scattered binary watermark. In this method SVM parameters are optimized by a modified optimal foraging algorithm, so that classification error can be reduced. Pixel rearrangement of the watermark image and host image makes the proposed method more secure, and it is highly robust as SVM is trained to classify locations that are less distorted by noise. Experimental outcomes depicts that the proposed method is accurate and better to the current cutting-edge methods in terms of security and robustness of digital image watermarking, as PSNR is approx. 72db and NC values are 1 after applying all the possible attacks.
{"title":"A highly robust and secure digital image watermarking using modified optimal foraging algorithm - based SVM classifier and hyperchaotic Fibonacci Q-matrix","authors":"Megha Gupta, R. Rama Kishore","doi":"10.1080/09720529.2022.2133247","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133247","url":null,"abstract":"Abstract Transmission media is at rise over the internet, due to which the copyright risk is alarming. In such situation, enhanced security is the main concern. Digital image watermarking is the technique that ensures the copyright protection, security, and authenticity of the digital image. This research work recommends a highly secured and robust digital image watermarking system. In this method, the cover image is confused by arbitrarily generated numbers by a six-dimensional hyperchaotic technique, then this permuted image is scattered by applying the Fibonacci Q matrix to generate a scattered host image. This scattered image is decomposed up to 3 levels through DWT and the low pass sub-band caused by DWT are further decomposed by SVD. Singular values generated by SVD are used for watermark embedding as a slight change in the value does not affect the image quality. SVM classifier is used to classify the appropriate location to insert the scattered binary watermark. In this method SVM parameters are optimized by a modified optimal foraging algorithm, so that classification error can be reduced. Pixel rearrangement of the watermark image and host image makes the proposed method more secure, and it is highly robust as SVM is trained to classify locations that are less distorted by noise. Experimental outcomes depicts that the proposed method is accurate and better to the current cutting-edge methods in terms of security and robustness of digital image watermarking, as PSNR is approx. 72db and NC values are 1 after applying all the possible attacks.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47226554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133239
K. Amrutha, P. Prabu
Abstract Natural Language Processing plays a vital role in our day-to-day life. Deep learning models for NLP help make human life easier as computers can think, talk, and interact like humans. Applications of the NLP models can be seen in many domains, especially in machine translation and psychology. This paper briefly reviews the different transformer models and the advantages of using an Encoder-Decoder language translator model. The article focuses on the need for sequence-to-sequence language-translation models like BERT, RoBERTa, and XLNet, along with their components.
{"title":"Effortless and beneficial processing of natural languages using transformers","authors":"K. Amrutha, P. Prabu","doi":"10.1080/09720529.2022.2133239","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133239","url":null,"abstract":"Abstract Natural Language Processing plays a vital role in our day-to-day life. Deep learning models for NLP help make human life easier as computers can think, talk, and interact like humans. Applications of the NLP models can be seen in many domains, especially in machine translation and psychology. This paper briefly reviews the different transformer models and the advantages of using an Encoder-Decoder language translator model. The article focuses on the need for sequence-to-sequence language-translation models like BERT, RoBERTa, and XLNet, along with their components.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42031743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133245
Tameem Ahmad, Maksud Ahamad, Sayyed Usman Ahmed, Nesar Ahmad
Abstract Evaluation of short answers is a challenging task. As there could be more than one way of expressing the same thing in a sentence by quite different words and phrases, evaluation through computer-based system of Short answers requires natural language understanding. Study has performed comparative analysis for short answer assessment with regression algorithms namely: Support Vector Regression, Linear Regression, Bagging Tree, Boosting Tree, Multilayer Perceptron Regressor, and Random Forest on extracted features. It proposes the combined features that take account of lexical, approximate string matching, and semantic similarity features. An empirical evaluation of feature selection is also done that further improves the results. These combined features achieved improved results as 0.71 & 0.78 for correlation and RMSE values respectively.
{"title":"Short question-answers assessment using lexical and semantic similarity based features","authors":"Tameem Ahmad, Maksud Ahamad, Sayyed Usman Ahmed, Nesar Ahmad","doi":"10.1080/09720529.2022.2133245","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133245","url":null,"abstract":"Abstract Evaluation of short answers is a challenging task. As there could be more than one way of expressing the same thing in a sentence by quite different words and phrases, evaluation through computer-based system of Short answers requires natural language understanding. Study has performed comparative analysis for short answer assessment with regression algorithms namely: Support Vector Regression, Linear Regression, Bagging Tree, Boosting Tree, Multilayer Perceptron Regressor, and Random Forest on extracted features. It proposes the combined features that take account of lexical, approximate string matching, and semantic similarity features. An empirical evaluation of feature selection is also done that further improves the results. These combined features achieved improved results as 0.71 & 0.78 for correlation and RMSE values respectively.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48470392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133246
Toyeer-E-Ferdoush, Habibur Rahman, M. Hasan
Abstract Sharing information from one device to another is gradually replacing hand-to-hand paper work in this connected digital age. Digital, modern technology are used to control the data communication. Because of this, the pace of a device’s cyber security is presently fast increasing. DDoS(Distributed Denial-of-Service), is one such phenomenon. TCP (Transmission Control Protocol) Half-open attacks include an SYN(Synchronization) flood attacks. It is a form of distributed denial of service attack that seeks to block all valid communication to a server in order to available server resources. This paper aims to protect the communication from DDoS TCP SYN flood attack. There are many research papers which can detect the attack after the attack take place and the prevention percentage is low. In this research paper this attack can be prevented much well than other model because a flood attack can detect before hampering the server and deny the connection attempt. There will be two cases studied and solved here that SYN-ACK(Synchronization-Acknowledgement) lost (no destination), SYN-ACK—no response.
摘要在这个互联的数字时代，从一台设备到另一台设备的信息共享正在逐渐取代手工的纸质工作。数字、现代技术被用来控制数据通信。正因为如此，设备网络安全的步伐目前正在快速增长。DDoS（Distributed Denial of Service，分布式拒绝服务）就是这样一种现象。TCP（传输控制协议）半开放攻击包括SYN（同步）洪水攻击。它是一种分布式拒绝服务攻击，旨在阻止与服务器的所有有效通信，以获得可用的服务器资源。本文旨在保护通信免受DDoS TCP SYN洪水攻击。有许多研究论文可以在攻击发生后检测到攻击，并且预防率很低。在这篇研究论文中，与其他模型相比，这种攻击可以很好地预防，因为洪水攻击可以在阻碍服务器之前进行检测并拒绝连接尝试。这里将研究并解决两种情况，即SYN-ACK（同步确认）丢失（无目的地），SYN-ACK——无响应。
{"title":"A convenient way to mitigate DDoS TCP SYN flood attack","authors":"Toyeer-E-Ferdoush, Habibur Rahman, M. Hasan","doi":"10.1080/09720529.2022.2133246","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133246","url":null,"abstract":"Abstract Sharing information from one device to another is gradually replacing hand-to-hand paper work in this connected digital age. Digital, modern technology are used to control the data communication. Because of this, the pace of a device’s cyber security is presently fast increasing. DDoS(Distributed Denial-of-Service), is one such phenomenon. TCP (Transmission Control Protocol) Half-open attacks include an SYN(Synchronization) flood attacks. It is a form of distributed denial of service attack that seeks to block all valid communication to a server in order to available server resources. This paper aims to protect the communication from DDoS TCP SYN flood attack. There are many research papers which can detect the attack after the attack take place and the prevention percentage is low. In this research paper this attack can be prevented much well than other model because a flood attack can detect before hampering the server and deny the connection attempt. There will be two cases studied and solved here that SYN-ACK(Synchronization-Acknowledgement) lost (no destination), SYN-ACK—no response.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45175942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133238
Soumi De, P. Prabu
Abstract Churn prediction is an active topic for research and machine learning approaches have made significant contributions in this domain. Models built to address customer churn, aim to identify customers who are at a high risk of terminating services offered by a company. Hence, an effective machine learning model indirectly contributes to the revenue growth of an organization, by identifying “at risk” customers, well in advance. This improves the success rate of retention campaigns and reduces costs associated with churn. The aim of this study is to explore the state-of-the-art machine learning techniques used in churn prediction. A systematic literature review, that is driven by 5 research questions and rigorous quality assessment criteria, is presented. There are 38 primary studies that are selected out of 420 studies published between 2018 and 2021. The review identifies popular machine learning techniques used in churn prediction and provides directions for future research. Firstly, the study finds that churn models lack generalization capability across industry domains. Hence, it identifies a need for researchers to explore techniques that extend beyond model experimentation, to improve efficiency of classifiers across domains. Secondly, it is observed that the traditional approaches to churn prediction depend significantly on demographic, product-usage, and revenue features alone. However, recent papers have integrated social network analysis-related features in churn models and achieved satisfactory results. Furthermore, there is a lack of scientific work that utilizes information-rich content of customer-company-interaction instances via email, chat conversations and other means. This area is the least explored. Thirdly, there is scope to investigate the effect of hybrid sampling strategies on model performance. This has not been extensively evaluated in literature. Lastly, there is no formal guideline on correct evaluation parameters to be used for models applied on imbalanced churn datasets. This is a grey area that requires greater attention.
{"title":"Predicting customer churn: A systematic literature review","authors":"Soumi De, P. Prabu","doi":"10.1080/09720529.2022.2133238","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133238","url":null,"abstract":"Abstract Churn prediction is an active topic for research and machine learning approaches have made significant contributions in this domain. Models built to address customer churn, aim to identify customers who are at a high risk of terminating services offered by a company. Hence, an effective machine learning model indirectly contributes to the revenue growth of an organization, by identifying “at risk” customers, well in advance. This improves the success rate of retention campaigns and reduces costs associated with churn. The aim of this study is to explore the state-of-the-art machine learning techniques used in churn prediction. A systematic literature review, that is driven by 5 research questions and rigorous quality assessment criteria, is presented. There are 38 primary studies that are selected out of 420 studies published between 2018 and 2021. The review identifies popular machine learning techniques used in churn prediction and provides directions for future research. Firstly, the study finds that churn models lack generalization capability across industry domains. Hence, it identifies a need for researchers to explore techniques that extend beyond model experimentation, to improve efficiency of classifiers across domains. Secondly, it is observed that the traditional approaches to churn prediction depend significantly on demographic, product-usage, and revenue features alone. However, recent papers have integrated social network analysis-related features in churn models and achieved satisfactory results. Furthermore, there is a lack of scientific work that utilizes information-rich content of customer-company-interaction instances via email, chat conversations and other means. This area is the least explored. Thirdly, there is scope to investigate the effect of hybrid sampling strategies on model performance. This has not been extensively evaluated in literature. Lastly, there is no formal guideline on correct evaluation parameters to be used for models applied on imbalanced churn datasets. This is a grey area that requires greater attention.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45530621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133242
S. Dhanda, Brahmjit Singh, P. Jindal, D. Panwar
Abstract With nearly 500 billion connected devices in 2025, information security will be the main concern of the researchers. It is the driving force in developing resource efficient cryptographic solutions. In this paper, we present a high throughput AES design with 32-bit data path that achieves the high efficiency via FPGA implementation. With the help of data path compression and effective utilization of FPGA architecture, the resource consumption is minimized. Galois field arithmetic is utilized for s-box implementation. Separate S-box for key generation has been employed to achieve higher throughput and low latency. The proposed design has been synthesized by PlanAhead software and implemented on different Xilinx FPGAs. It is compared with AES implementations. With a throughput of 2.34 Gbps and efficiency of 5.10 Mbps/slice, the design outperforms different lightweight ciphers. High throughput and low latency make it suitable for surveillance applications in IoT and smart grid.
{"title":"A highly efficient FPGA implementation of AES for high throughput IoT applications","authors":"S. Dhanda, Brahmjit Singh, P. Jindal, D. Panwar","doi":"10.1080/09720529.2022.2133242","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133242","url":null,"abstract":"Abstract With nearly 500 billion connected devices in 2025, information security will be the main concern of the researchers. It is the driving force in developing resource efficient cryptographic solutions. In this paper, we present a high throughput AES design with 32-bit data path that achieves the high efficiency via FPGA implementation. With the help of data path compression and effective utilization of FPGA architecture, the resource consumption is minimized. Galois field arithmetic is utilized for s-box implementation. Separate S-box for key generation has been employed to achieve higher throughput and low latency. The proposed design has been synthesized by PlanAhead software and implemented on different Xilinx FPGAs. It is compared with AES implementations. With a throughput of 2.34 Gbps and efficiency of 5.10 Mbps/slice, the design outperforms different lightweight ciphers. High throughput and low latency make it suitable for surveillance applications in IoT and smart grid.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41824573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133252
S. Maheshwari, S. Shiwani, S. Choudhary
Abstract Cloud computing aims to optimal use of its resources by aggregating them to increase throughput and solve difficult computational problems in the most efficient way possible. Task scheduling problem is incompliant with exact solutions in cloud due to its NP-hard nature. To address this, various meta-heuristic strategies have been developed. A task scheduler should locate the optimal resources for the user’s job while taking into account specific cloud task parameter constraints. Here, a hybrid task scheduling strategy is described that incorporates deep learning and nature-inspired meta-heuristic optimization to maximize cloud throughput while minimizing completion time in an IaaS cloud. The scheduler succeeds towards cloudlet allocation resulted to shorter makespan and higher system throughput. The novel scheduling technique was evaluated against certain algorithms using the CloudSim software. When compared to existing algorithms such as Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO), the experimental findings show that the suggested approach outperforms them.
{"title":"Optimizing scheduling in cloud using a meta-heuristic approach","authors":"S. Maheshwari, S. Shiwani, S. Choudhary","doi":"10.1080/09720529.2022.2133252","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133252","url":null,"abstract":"Abstract Cloud computing aims to optimal use of its resources by aggregating them to increase throughput and solve difficult computational problems in the most efficient way possible. Task scheduling problem is incompliant with exact solutions in cloud due to its NP-hard nature. To address this, various meta-heuristic strategies have been developed. A task scheduler should locate the optimal resources for the user’s job while taking into account specific cloud task parameter constraints. Here, a hybrid task scheduling strategy is described that incorporates deep learning and nature-inspired meta-heuristic optimization to maximize cloud throughput while minimizing completion time in an IaaS cloud. The scheduler succeeds towards cloudlet allocation resulted to shorter makespan and higher system throughput. The novel scheduling technique was evaluated against certain algorithms using the CloudSim software. When compared to existing algorithms such as Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO), the experimental findings show that the suggested approach outperforms them.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43309317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133243
K. Manjunatha, M. Kempanna
Abstract A web application is a dynamic, intricate, and interactive program that provides end-users with information and services such as utility payments, online communication, e-learning, socializing, shopping, online banking, and income tax filing etc. Web applications have become a major target for attackers due to their accessibility, availability, and ubiquity. Web application vulnerabilities are hazardous for some reasons. Attackers can harm an organizations image and status. The implementation flaws in web application allow the invader to infuse user-input that violates the syntax-based assembly of the query or infuse malicious code etc. Among various types of injection flaws, SQL injection (SQLI) is more prominent than (XML) both are considered as common application-layer web attack, which allows the attacker to bypass the security mechanisms therefore; these two are ranked as the most common vulnerabilities. Hence, a methodology for detecting evaluating both SQLI & XML vulnerabilities in web applications are considered for research. This research work addresses the above mentioned flaws and proposed an Ensemble Method to classify the Structure Query Language injection vulnerabilities, we selected a benchmark dataset with 33,758 rows containing; various types of SQL and XML injection attacks. Raw data is preprocessed to remove artifacts, and then feature engineering is performed using Natural Language Processing techniques to clean the data and extract 6 types of features such as TF-IDF, Word-to-Vector, SkipGram, Count Vectorizer, Glove and Continuous Bag of words. Imbalance data is handled using sampling techniques, best features are selected using 4 types of validation techniques Significant Test, PCA, Variance Threshold and Sbest. Prepared data is provided to Ensemble Model having two stages; Stage-2 accepts URL from the user and detects presence of susceptibility in the sub domains and domains. Stage-1 having 9 different types of machine learning models Multinomial, Gaussian, Bernoulli Naive Bayes, Logistic Regression, Decision Tree, Random Forest, AdaBoost, SVC with, poly, rbf and linear kernel, these models are trained on additional vectors such as google news and glove to detect the new query either SQL or XML for presences or absence of vulnerability, using this proposed ensemble approach obtained the accuracy of 99%.
摘要web应用程序是一个动态、复杂和交互式的程序，它为最终用户提供信息和服务，如公用事业支付、在线通信、电子学习、社交、购物、网上银行和所得税申报等。由于其可访问性、可用性和普遍性，web应用程序已成为攻击者的主要目标。由于某些原因，Web应用程序漏洞是危险的。攻击者可能会损害组织的形象和地位。web应用程序中的实现缺陷允许入侵者注入违反查询基于语法的汇编的用户输入或注入恶意代码等。在各种类型的注入缺陷中，SQL注入（SQLI）比XML更突出，两者都被认为是常见的应用层web攻击，这使攻击者能够绕过安全机制；这两个漏洞被列为最常见的漏洞。因此，需要考虑一种检测和评估web应用程序中SQLI和XML漏洞的方法进行研究。本研究工作针对上述缺陷，提出了一种集成方法来对结构查询语言注入漏洞进行分类，我们选择了一个33758行的基准数据集；各种类型的SQL和XML注入攻击。对原始数据进行预处理以去除伪影，然后使用自然语言处理技术进行特征工程以清理数据并提取6种类型的特征，如TF-IDF、Word to Vector、SkippGram、Count Vectorizer、Glove和Continuous Bag of words。不平衡数据使用采样技术处理，最佳特征使用4种类型的验证技术显著性检验、主成分分析、方差阈值和Sbest进行选择。准备好的数据被提供给具有两个阶段的集合模型；阶段2接受来自用户的URL，并检测子域和域中是否存在易感性。阶段-1具有9种不同类型的机器学习模型多项式、高斯、伯努利-奈夫贝叶斯、逻辑回归、决策树、随机森林、AdaBoost、SVC，以及poly、rbf和线性内核，这些模型在谷歌新闻和手套等附加向量上进行训练，以检测SQL或XML的新查询是否存在漏洞，使用该集成方法获得了99%的准确率。
{"title":"Count vectorizer model based web application vulnerability detection using artificial intelligence approach","authors":"K. Manjunatha, M. Kempanna","doi":"10.1080/09720529.2022.2133243","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133243","url":null,"abstract":"Abstract A web application is a dynamic, intricate, and interactive program that provides end-users with information and services such as utility payments, online communication, e-learning, socializing, shopping, online banking, and income tax filing etc. Web applications have become a major target for attackers due to their accessibility, availability, and ubiquity. Web application vulnerabilities are hazardous for some reasons. Attackers can harm an organizations image and status. The implementation flaws in web application allow the invader to infuse user-input that violates the syntax-based assembly of the query or infuse malicious code etc. Among various types of injection flaws, SQL injection (SQLI) is more prominent than (XML) both are considered as common application-layer web attack, which allows the attacker to bypass the security mechanisms therefore; these two are ranked as the most common vulnerabilities. Hence, a methodology for detecting evaluating both SQLI & XML vulnerabilities in web applications are considered for research. This research work addresses the above mentioned flaws and proposed an Ensemble Method to classify the Structure Query Language injection vulnerabilities, we selected a benchmark dataset with 33,758 rows containing; various types of SQL and XML injection attacks. Raw data is preprocessed to remove artifacts, and then feature engineering is performed using Natural Language Processing techniques to clean the data and extract 6 types of features such as TF-IDF, Word-to-Vector, SkipGram, Count Vectorizer, Glove and Continuous Bag of words. Imbalance data is handled using sampling techniques, best features are selected using 4 types of validation techniques Significant Test, PCA, Variance Threshold and Sbest. Prepared data is provided to Ensemble Model having two stages; Stage-2 accepts URL from the user and detects presence of susceptibility in the sub domains and domains. Stage-1 having 9 different types of machine learning models Multinomial, Gaussian, Bernoulli Naive Bayes, Logistic Regression, Decision Tree, Random Forest, AdaBoost, SVC with, poly, rbf and linear kernel, these models are trained on additional vectors such as google news and glove to detect the new query either SQL or XML for presences or absence of vulnerability, using this proposed ensemble approach obtained the accuracy of 99%.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48488623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133249
Usama Bin Rashidullah Khan, N. Akhtar, Umar Tahir Kidwai, Ghufran Alam Siddiqui
Abstract Business and brand owners are using social media networks to provide and deliver various services to their clients and collect information about their products from customers. Customers give their opinions as well as ideas for the improvement of the products on the review platforms and portals. Suggestion Mining is a technique of automatic extraction of these innovative ideas or suggestions from online source data. In this paper, we proposed TCN architecture for suggestion mining from online reviews. The TCN uses causal and dilated convolutional layers to process sequential or temporal data and captures long-term dependencies. TCN architecture on the dataset of SemEval-2019 subtask A is experimented. The dataset is highly imbalanced and to overcome this problem, the ensemble oversampling technique to balance the dataset is applied. TCN is also experimented with the attention mechanism. Our proposed model outperforms the existing works by achieving an F1 score of 82.0 %.
{"title":"Suggestion mining from online reviews using temporal convolutional network","authors":"Usama Bin Rashidullah Khan, N. Akhtar, Umar Tahir Kidwai, Ghufran Alam Siddiqui","doi":"10.1080/09720529.2022.2133249","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133249","url":null,"abstract":"Abstract Business and brand owners are using social media networks to provide and deliver various services to their clients and collect information about their products from customers. Customers give their opinions as well as ideas for the improvement of the products on the review platforms and portals. Suggestion Mining is a technique of automatic extraction of these innovative ideas or suggestions from online source data. In this paper, we proposed TCN architecture for suggestion mining from online reviews. The TCN uses causal and dilated convolutional layers to process sequential or temporal data and captures long-term dependencies. TCN architecture on the dataset of SemEval-2019 subtask A is experimented. The dataset is highly imbalanced and to overcome this problem, the ensemble oversampling technique to balance the dataset is applied. TCN is also experimented with the attention mechanism. Our proposed model outperforms the existing works by achieving an F1 score of 82.0 %.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42286487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-03DOI: 10.1080/09720529.2022.2133248
Md Mehedi Hasan Real, Anamika Zaman Priya, Md Alomgir Hossain, Khandakar Rabbi Ahmed
Abstract In this day and age, the programmed procurement of a tag and acknowledgment assumes a significant part in observing and coordinating vehicles in significant urban communities. It is hard to recognize a driver or proprietor of a vehicle that disregards traffic controls or plays out any incidental movement out and about. It will require a great deal of investment for a cop to review the plate of every vehicle. Subsequently, a mechanized tag acknowledgment framework can tackle these sorts of issues. This is the proposed technique; two Deep Learning calculations are utilized to distinguish the permit number and characters on the tag from the constant picture. The primary YOLOv5 model tracks down the main in the live video of a vehicle out and about. Then, at that point, cut out the area of the permit numbers in the video. The cut casing is then embedded into a second SSD (Single Shot Detection) to identify slugs on that tag. The prepared model acquires a high precision of 96.2% over a sum of 400 picture databases.
{"title":"A sustainable competing dynamic - Real-time Bangla license plate detection and recognition system using YOLOv5 and SSD: A deep learning application","authors":"Md Mehedi Hasan Real, Anamika Zaman Priya, Md Alomgir Hossain, Khandakar Rabbi Ahmed","doi":"10.1080/09720529.2022.2133248","DOIUrl":"https://doi.org/10.1080/09720529.2022.2133248","url":null,"abstract":"Abstract In this day and age, the programmed procurement of a tag and acknowledgment assumes a significant part in observing and coordinating vehicles in significant urban communities. It is hard to recognize a driver or proprietor of a vehicle that disregards traffic controls or plays out any incidental movement out and about. It will require a great deal of investment for a cop to review the plate of every vehicle. Subsequently, a mechanized tag acknowledgment framework can tackle these sorts of issues. This is the proposed technique; two Deep Learning calculations are utilized to distinguish the permit number and characters on the tag from the constant picture. The primary YOLOv5 model tracks down the main in the live video of a vehicle out and about. Then, at that point, cut out the area of the permit numbers in the video. The cut casing is then embedded into a second SSD (Single Shot Detection) to identify slugs on that tag. The prepared model acquires a high precision of 96.2% over a sum of 400 picture databases.","PeriodicalId":46563,"journal":{"name":"JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY","volume":null,"pages":null},"PeriodicalIF":1.4,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44086446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：